PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address.

Keryn Headlines

Weekly, week 52 of 2025 · Retrospective · Covers 16 December 2025 12:00 to 23 December 2025 12:00 UTC

Read on the web

Top developments

Sonicwall warns of new SMA1000 zero-day exploited in attacks

BleepingComputer · Incident · Remediation

New critical WatchGuard Firebox firewall flaw exploited in attacks

BleepingComputer · Incident · Remediation

Malicious npm package steals WhatsApp accounts and messages

BleepingComputer · Analysis · SCA

Continuously hardening ChatGPT Atlas against prompt injection

OpenAI News · Research · AI Security, Remediation

Catching Malicious Package Releases Using a Transparency Log

OpenSSF · Incident · SCA, Trust

SAST

Multi-LLM Energy Reasoning for Binary-Free Zero-Day IoT Detection

arXiv · Research · DAST, SAST

You can now require reviews before closing Dependabot alerts with delegated alert dismissal

GitHub · Analysis · SAST, Trust

What You Trust Is Insecure: Demystifying How Developers (Mis)Use Trusted Execution Environments in Practice

arXiv · Research · SAST

A Systematic Study of Code Obfuscation Against LLM-based Vulnerability Detection

arXiv · Research · SAST

SCA

Most Parked Domains Now Serving Malicious Content

Krebs on Security · Research · SCA

CISA flags ASUS Live Update CVE, but the attack is years old

BleepingComputer · Analysis · SCA

A Practical Solution to Systematically Monitor Inconsistencies in SBOM-based Vulnerability Scanners

arXiv · Research · SCA

Insecure Ingredients? Exploring Dependency Update Patterns of Bundled JavaScript Packages on the Web

arXiv · Research · SCA

DAST

From Coverage to Causes: Data-Centric Fuzzing for JavaScript Engines

arXiv · Research · DAST

Attention Distance: A Novel Metric for Directed Fuzzing with Large Language Models

arXiv · Research · DAST

WuppieFuzz: Coverage-Guided, Stateful REST API Fuzzing

arXiv · Research · DAST

Penetration Testing of Agentic AI: A Comparative Security Analysis Across Models and Frameworks

arXiv · Research · AI Security, DAST

Agentic SDLC

CVE-2025-14733: WatchGuard Firebox Out of Bounds Write Vulnerability added to CISA's Known Exploited Vulnerabilities catalog

CISA · Advisory · Agentic SDLC

Reflection-Driven Control for Trustworthy Code Agents

arXiv · Research · Agentic SDLC

SecureCode: A Production-Grade Multi-Turn Dataset for Training Security-Aware Code Generation Models

arXiv · Research · Agentic SDLC, AI Security

SoK: Understanding (New) Security Issues Across AI4Code Use Cases

arXiv · Research · Agentic SDLC, Models

AI-BOM

Evolving AI Transparency: The Journey of the AIBOM Generator and Its New Home at OWASP

OWASP GenAI Security Project · Analysis · AI-BOM

Remediation

HPE warns of maximum severity RCE flaw in OneView software

BleepingComputer · Analysis · Remediation

Hackers exploit newly patched Fortinet auth bypass flaws

BleepingComputer · Advisory · Remediation

Well Begun is Half Done: Location-Aware and Trace-Guided Iterative Automated Vulnerability Repair

arXiv · Research · Remediation

Toward Training Superintelligent Software Agents through Self-Play SWE-RL

arXiv · Research · Remediation

Trust

CRA Implementation Resources from the European Commission

OpenSSF · Standard / policy · Trust

MathLedger: A Verifiable Learning Substrate with Ledger-Attested Feedback

arXiv · Research · Trust

Computable Gap Assessment of Artificial Intelligence Governance in Children's Centres: Evidence-Mechanism-Governance-Indicator Modelling of UNICEF's Guidance on AI and Children 3.0 Based on the Graph-GAP Framework

arXiv · Research · Trust

How frontier AI companies could implement an internal audit function

arXiv · Research · Trust

Models

Optimistic TEE-Rollups: A Hybrid Architecture for Scalable and Verifiable Generative AI Inference on Blockchain

arXiv · Research · Models

Odysseus: Jailbreaking Commercial Multimodal LLM-integrated Systems via Dual Steganography

arXiv · Research · Models

Conditional Adversarial Fragility in Financial Machine Learning under Macroeconomic Stress

arXiv · Research · Models

A Multi-Perspective Benchmark and Moderation Model for Evaluating Safety and Adversarial Robustness

arXiv · Research · Models

AI Security

From MCP to Vibe Coding: Full Endpoint Visibility in Wiz AI Security

Wiz · Analysis · AI Security

DASH: Deception-Augmented Shared Mental Model for a Human-Machine Teaming System

arXiv · Research · AI Security

Securing Agentic AI Systems -- A Multilayer Security Framework

arXiv · Research · AI Security

Beyond the Benchmark: Innovative Defenses Against Prompt Injection Attacks

arXiv · Research · AI Security

Advisories in this window

  • 102 npm advisories · 90 critical · 8 high · 1 moderate · 3 low
  • 19 PyPI advisories · 2 critical · 4 high · 10 moderate · 3 low
  • 15 Go advisories · 1 critical · 7 high · 5 moderate · 2 low
  • 9 Packagist advisories · 2 critical · 1 high · 6 moderate
  • 7 Maven advisories · 2 high · 5 moderate
  • 5 NuGet advisories · 3 moderate · 2 low
  • 1 RubyGems advisory · 1 moderate

All advisories

Titles, links and short excerpts only; every item links to its source. Selection is automatic; the editor's note is AI-drafted and checked as described. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.